Stochastic Reconstruction of Sentinel-1 Wind Speed Image Time Series via Multiple-Point Geostatistical Simulation
Date Issued
November 2023
Author(s)
Advisor
Abstract
Offshore wind is expected to play a key role in future energy systems. Wind energy assessment studies often call for long-term and spatially consistent datasets. Despite the vast amount of available data sources, no current means can provide relevant sub-daily information at a fine spatial scale (~1km). Synthetic Aperture Radars (SAR) deliver wind field estimates over the ocean at fine spatial resolution but suffer from partial coverage and irregular revisit times. Numerical Weather Prediction (NWP) models, which are the basis of reanalysis products, can be queried at any time step but lack fine scale spatial variability. In this dissertation, a geostatistical methodology is developed to combine the advantages of both under the framework of Multiple-Point Statistics and is employed to realistically reconstruct wind speed patterns at time instances where satellite information is absent. Synthetic fine-resolution wind speed images are generated conditioned to co-registered coarse-scale regional reanalysis information. Available simultaneous data sources are used as training data to generate the synthetic image time series. The latter are then evaluated via cross-validation and statistical comparison against reference satellite data. Multiple realizations are also generated to assess the uncertainty associated with the simulation outputs. Overall, results demonstrate the effectiveness of the proposed method in realistically reconstructing the fine-scale wind speed spatiotemporal variability over the offshore area of Cyprus and can provide the basis for an offshore wind resource assessment. The proposed framework can be generalized to other regions, scales or variables.
File(s)![Thumbnail Image]()
Name
PhD_StylianosHadjipetrou_Final_Signed.pdf
Size
13.38 MB
Format
Adobe PDF
Checksum (MD5)
1401525d0a5aac8ab937433fa28fff07